The Institute of Visual Computing at TU Graz, is thrilled to share a tremendous milestone for our research community and for the university: our head, Thomas Pock, has been awarded an ERC Advanced Grant for his project EAGLE – Efficient Algorithms for Generative Learning. This is the first ERC Advanced Grant at TU Graz and provides 2.5 million euros in funding for supporting excellent frontier research.
With EAGLE, Thomas aims to advance computer vision with novel generative learning methods and algorithms. A central focus of EAGLE is on new mathematical methods for Bayesian inverse problems. Instead of deriving a single result from the measurement data, his approach will generate many different, yet equally plausible solutions. This method employs generative AI and novel, efficient sampling algorithms, showing which image details are data-supported and where uncertainties remain (e.g., in MRI).
EAGLE tightly links data analysis with data generation: models not only interpret data but also synthesize realistic examples, improving reliability and uncertainty quantification in medical imaging and beyond. Importantly, EAGLE takes a different path from the trend toward ever-larger, compute-heavy AI models. “The aim is to show that significant progress can be made even with a moderate use of data and computing power – based on sound mathematical theory,” says Thomas.
We warmly congratulate Thomas on this outstanding achievement and wish him best of luck with his research!
Read the full press release here.
